© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36073
    Prev
    Next
    INDEX
    Explanations

    hard-coded, into, build

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ula
    -0.10
    uta
    -0.09
    ä¸įåıĹ
    -0.09
    upil
    -0.09
    Decor
    -0.08
    ulta
    -0.08
    åħ±åIJĮåĬªåĬĽ
    -0.08
    è¹
    -0.08
     Cook
    -0.08
     Cooking
    -0.08
    POSITIVE LOGITS
    è¿Ľåİ»
    0.19
     into
    0.17
    è¿ĽæĿ¥
    0.16
    纳åħ¥
    0.14
    Into
    0.14
    èŀįåħ¥
    0.14
     INTO
    0.13
     vÃło
    0.13
    _into
    0.13
    åµĮ
    0.13
    Activations Density 0.116%

    No Known Activations